# How to set one checkpoint per agent in a multiagent config?

**URL:** <https://discuss.ray.io/t/how-to-set-one-checkpoint-per-agent-in-a-multiagent-config/6596>\
**Category:** RLlib\
**Created:** [June 22, 2022, 1:57pm UTC](https://discuss.ray.io/t/how-to-set-one-checkpoint-per-agent-in-a-multiagent-config/6596 "2022-06-22T13:57:41Z")\
**Posts on this page:** 1\
**Showing post:** 2

<div class="post-metadata">

**Author:** ![2dm](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/2dm/32/885_2.png) [@2dm](https://discuss.ray.io/u/2dm)\
**Post date:** [June 22, 2022, 4:46pm UTC](https://discuss.ray.io/t/how-to-set-one-checkpoint-per-agent-in-a-multiagent-config/6596/2 "2022-06-22T16:46:07Z")

</div>

As far as I know, there is still no option to do that, but you can workaround that though.  
If you use the same environment, you can just load your checkpoint and only compute actions to the selected agent.  
If not, you will need to load the checkpoint in the environment you trained it and then to save the selected agent’s _weights_, which you will be able to load later.  
You can see a code for example [here](https://discuss.ray.io/t/ppo-load-checkpoint-from-previous-version-fails/5307/3).

Keep in mind that you’re saving only the weights and not the entire state like in a checkpoint.

---

_[View the full topic](https://discuss.ray.io/t/how-to-set-one-checkpoint-per-agent-in-a-multiagent-config/6596)._